See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/326112777

Mechanical properties and bond strength degradation of GFRP and steel

rebars at elevated temperatures

Article  in  Construction and Building Materials · July 2018

DOI: 10.1016/j.conbuildmat.2018.06.203

CITATIONS READS

29 769

4 authors:

Fatih Mehmet ÖZKAL Mehmet Polat

Ataturk University Ataturk University
27 PUBLICATIONS   100 CITATIONS    2 PUBLICATIONS   29 CITATIONS   

SEE PROFILE SEE PROFILE

Mahmud Yağan Muhammed Orhan Öztürk

2 PUBLICATIONS   29 CITATIONS    2 PUBLICATIONS   29 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Elevated temperature effects on concrete members reinforced with different materials View project

GFRP ve çelik donatıların yüksek sıcaklık etkileri altında aderans kayıplarının incelenmesi View project

All content following this page was uploaded by Fatih Mehmet ÖZKAL on 02 July 2018.

The user has requested enhancement of the downloaded file.

Mechanical properties and bond strength
degradation of GFRP and steel rebars at
elevated temperatures
Özkal FM, Polat M, Yağan M, Öztürk MO (2018). Mechanical properties and bond strength
degradation of GFRP and steel rebars at elevated temperatures. Construction and Building
Materials, 184, 45-57.
https://doi.org/10.1016/j.conbuildmat.2018.06.203

F
Fatih Mehmet ÖZKAL1,*,
Mehmet POLAT2, Mahmud YAĞAN2, Muhammed Orhan ÖZTÜRK2

OO
(*Corresponding Author)
1 Department of Civil Engineering, Atatürk University, Erzurum 25240, Turkey
2 Department of Civil Engineering, Erzincan University, Erzincan 24060, Turkey
PR
ABSTRACT
OR

Fiber reinforced polymer (FRP) rebars have recently been gaining popularity in the construction
industry all around the world while steel rebars have been widely used so far. FRP bars, which
have higher tensile strength compared to steel rebars with the same nominal diameter under
TH

normal conditions, are composed of resin matrix and fibers. In this study, mechanical and
bonding properties of glass fiber reinforced polymer (GFRP) and steel rebars subjected to
elevated temperature effects were investigated throughout a correlative comparison. Axial
tensile tests and pull-out tests of these materials were performed subsequent to exposing them
AU

into elevated temperature effects in the range of 23-800°C. Severe effects on the tensile
properties of bare steel bars were observed after 600°C, while this critical limit is 300°C for
bare GFRP bars. Test results show that bond strength degradation is almost linear for both type
of rebars, however, 600°C is also the critical temperature regarding the serious deterioration on
concrete. Additionally, an empirical modeling approach on the bond strength degradation of
rebars at elevated temperatures was proposed and it was found that comparison results have
quite promising consistency based on the modeling process.

Keywords: fire effect; elevated temperatures; bond strength; glass fiber reinforced polymers;
GFRP rebars; tensile tests, pull-out tests.

1
1. INTRODUCTION

High temperature levels, when considering RC structural elements in the existing designs,
affect initially the concrete and then the reinforcement bars. Appropriate design of each
concrete component could enhance the concrete behavior against elevated temperature effects.
For instance; when silica based aggregates and cement are used in the concrete mortar, it is
observed that at 573°C, 15% of the silica's volumetric expansion results in the fragmentize of
the concrete (Yamazaki et al. 1995). A fresh concrete that is not correctly placed and does not
reach adequate hydration, contains free water and this water evaporates at 100°C, causing
fragmentation. In addition, this situation accelerates the evaporation of the bound water at

F
300°C in the hydrated elements (Bingöl and Gül 2009a; Handoo et al. 2002). Thus, rapid

OO
degradation of the concrete strength causes the temperature increase in reinforcement bars. On
the other hand, calcium hydroxide, which is an important cement component, shrinks by 33%
by losing water and transforms into quicklime at 530°C. During the fire, the water that is
squeezed into the structure causes the quicklime changes from calcium hydroxide to calcium
PR
hydroxide, resulting in a volume expansion of 44%, and this sudden volume conversion causes
cracking of the concrete (Khoury 1996). Reinforcement steel starts to lose its yield strength
after 600°C at these stages, alongside even physical damages (Ergün et al. 2009; Eurocode 2
Part 1-2 2004).
OR

Considering these situations, which will lead to the end of the service life of the building,
enhancement of material properties also could improve of the structural performance of
buildings. GFRP bars have recently been used in the construction industry for this purpose,
TH

especially considering their high tensile strength and durability in harsh environments, such as
corrosion effects. In other respects, having a low value of the glass transition temperature of the
polymeric matrix that can be even close to 80°C; GFRP bars exhibit inadequate resistance to
elevated temperatures. This level of temperature is higher than the thermal loads due to seasonal
AU

changes and solar radiation in bridge decks (Carvelli et al. 2013).

In this study, following a review summary of previous research about elevated temperature
effects and bonding behavior of concrete-reinforcement bars; mechanical properties and
bonding performance of GFRP and steel rebars subjected to elevated temperature effects were
investigated and evaluated considering their use in reinforced concrete structures. Therefore,
elevated temperature effects within the range of 23-800°C (23, 100, 150, 200, 250, 300, 400,
500, 600 and 800°C) were investigated throughout cubic compressive strength tests of concrete,
axial tensile tests of steel and GFRP rebars, and pull-out tests of steel and GFRP rebars. Finally,
an empirical model for the bond strength of rebars, which are subjected to elevated
temperatures, was developed based on the experimental results of this study.
2
2. REVIEW OF THE PREVIOUS RESEARCH

Gustaferro et al. (1971) performed an experimental research that could be regarded as one of
the preliminary research on the fire resistance of lightweight and isolating concrete. In this
study, fire resistance of concrete was examined as well as the relationship between the moisture
content of the samples and the relative humidity. As the most important outcome of the study,
it was shown that the increase in unit weight causes the decrease of fire resistance for each
concrete type. Zoldners and Wilson (1973) exposed concrete mixtures of expanded schist and
cinder aggregates to elevated temperature effects. As a result of this study, it was found that

F
full-light concrete shows better strength than semi-light concrete under different temperature

OO
effects; and the use of blast furnace slag does not provide an advantage against fire effects. On
the contrary, it affects the bending strength of concrete in the negative direction. Rostasy et al.
(1980) investigated the effect of elevated temperatures on the porous structure of the concrete
by means of mercury porosimetry method and found that elevated temperatures causes an
PR
increase in the total pore volume.

In order to go through the studies on bond strength behavior without elevated temperature
OR

effects, Larrard et al. (1993) investigated the effect of bar diameter and surface condition on the
bond strength between high-performance concrete and rebars. Experimental results revealed
that the use of high-performance concrete significantly increases the bond strength and this
increase in bond strength may be due to the increase of the tensile strength of the concrete.
TH

Furthermore, an increase at a higher rate was observed with the use of smaller reinforcement
regarding the scale effect. Benmokrane et al. (1996) compared the load-bearing capacity and
bond strength of GFRP anchor bars with steel bars. It was found that GFRP bars have lower
AU

bond strength values than steel bars. Within the study of Fu and Chung (1997), it was stated
that bond strength between concrete and rebar increases with increasing compressive strength
according to the previous studies, so bond strength is considered to decrease with increasing
water/cement ratio. On the contrary, it has been seen that increasing the water/cement ratio
enhances the bond strength, while the increase in relative humidity and the use of moisturized
reinforcement increase the bond strength. Tighiouart et al. (1998) compared the bond strength
of GFRP bars with steel bars and reported that the maximum bond strength value decreases
with the increase of the rebar diameter. It was presented that bond strength of GFRP bars are
lower than that of steel bars because the adhesion and friction characteristics of the rebar affect
the bond strength depending on the experimental results. Chang et al. (2002) investigated the

3
effects of different sizes of river sand and sand/epoxy percentage by weight in their study, which
was designed to enhance bond strength of epoxy-coated reinforcement bars. Lee et al. (2002)
conducted a pull-out testing process by using accelerated electrical corrosion method to
investigate the effect of corrosion on bonding properties of concrete and rebar. De Lorenzis et
al. (2002) studied to determine the mechanism of bonding between FRP bars and concrete and
to analyze the most effective parameters on bond strength. Variables such as the type of FRP
bar, the properties of the material filling the ribs, anchorage length and rib size, were
investigated. It was reported that fracture of concrete pull-out samples, in which FRP bars were
embed via epoxy, occurs at the epoxy-concrete interface, and average value of bond strength at

F
the epoxy-concrete interface decreases with increasing anchorage length and rib size. Gallego

OO
et al. (2004) demonstrated a comparison of bonding behavior between concrete and steel bars
for two different types, black steel and galvanized steel, using pull-out tests. Cheng et al. (2005)
investigated the corrosion resistance of zinc-coated rebars and the effect of coating on bond
strength between concrete and steel rebars. Banholzer et al. (2006) presented a numerical
PR
solution method to determine bonding characteristics over bond-slip relationship and confirmed
the results with the data obtained from the pull-out tests. Jendele and Cervenka (2006) also
suggested a numerical model for determining the bonding performance level between concrete
and rebars. Yavuz (2011) performed a detailed research on the use of fiber-reinforced polymers
OR

as rebars in RC beams. Detailed evaluations were made on the bonding condition of FRP bars in
concrete. In the study carried out by Cullazoğlu (2014), numerical analyzes of concrete beams
with FRP rebars, subjected to bending and shear effects, were performed. Within this study,
TH

which provides important information on the behavior of bending and shear with regard to the
use of steel and FRP bars, it was emphasized that combined use of two types of bars in hybrid
beams has the capacity of increases in both ductility and rigidity performance. Fava et al. (2016)
reported the results of an extensive investigation on the bonding behavior between GFRP bars
AU

and concrete. Pull-out tests were performed on helically wrapped and sand-coated GFRP rebars
with a wide range of diameters, as well as nonlinear finite element simulations.

It is important to know the effects of elevated temperatures on the bond strength between
concrete and rebar so that fire resistance and residual structural performance of the RC buildings
can be fully understood. As one of the pioneer studies considering elevated temperature effects
on construction materials, Morley and Royles (1980) studied the strength of concrete and steel,
under and after high temperature levels. Then the methods used to determine bond strength at
ambient temperature were listed, researches on bonding strength at high temperature levels
were evaluated and suggestions were presented regarding the structural performance.
4
Diederichs and Schneider (1981) investigated the bonding behavior of steel bars and concrete
at elevated temperatures. Although the most important parameter in terms of the performance
of reinforced concrete buildings under fire is bonding, it was stated that there is not enough
information about this problem. El-Hawary and Hamoush (1996) conducted an experimental
study to determine bonding behavior at the interface between concrete and rebars under
elevated temperature effects, and also the effects of temperature, heating duration and cooling
method on bonding performance. In the study of Katz et al. (1999), comparing the bonding
properties of FRP bars with ordinary deformed steel bars at elevated temperatures, 80-90%
reduction of FRP bond strength has been observed while the reduction of steel bond strength is

F
38% in the same temperature range. Greater sensitivity to high temperatures was found in FRP

OO
bars, in which the bond relies mainly on the polymer treatment at the surface of the bar. Katz
and Berman (2000) investigated the bonding behavior between FRP rebars and concrete in
addition to the proposal of a model for predicting the bond strength under elevated temperature
effects. Reduction in maximum anchorage force has been noted for all bars with increasing
PR
temperature. With increasing temperature, bond strength decrease was also detected in samples
using steel rebars. Chiang et al. (2000) conducted a series of combinational tests on sound
transmission and pull-out methods to determine the bonding between concrete and rebars after
fire damage, while Hashimoto and Takiguchi (2004) performed an experimental study to
OR

determine the behavior of anchor bars, during and after high temperatures effects. Haddad and
Abendeh (2004) investigated the effect of synthetic and brass-coated short steel fibers to
maintain the bonding between concrete and rebars exposed to the heating-cooling cycles.
TH

Similarly, Chen and Liu (2004) investigated the residual performance of high-strength and
fiber-reinforced high-strength concrete behavior under elevated temperature effects. It was
reported that the use of carbon and steel fiber delays the explosion time, and the polypropylene
fiber suppresses this phenomenon. It was also determined that residual compressive and tensile
AU

strength levels of fiber-reinforced high-strength concrete are higher than high-strength concrete
without fiber. Yuan et al. (2005) examined bonding performance in reinforced concrete after
different high temperatures and cooling methods.

The effects of water and alkali conditions on bonding between concrete and rebars, and stiffness
of GFRP bars at elevated temperatures, were investigated by Abbasi and Hogg (2005). It was
found that fiber-concrete interface deteriorates with increasing temperature. The change in
stress transfer between concrete and rebar was presented as a result of the decrease in bonding
alongside the increase of interfacial adherence over time at room temperature. An experimental
research was conducted by Wang and Kodur (2005) to investigate the strength and stiffness
5
properties of FRP bars at high temperature levels. Two different types of FRP bars, carbon fiber
(CFRP) and glass fiber (GFRP) were used. Critical temperatures for the decrease of tensile
strength for steel, GFRP and CFRP bars were presented. Lau and Anson (2006) studied the
compressive strength, flexural strength, modulus of elasticity and porosity of concrete with steel
fiber content after exposure to elevated temperatures. It was found that use of fiber improves
the mechanical properties of the concrete. Galati et al. (2006) analyzed the bonding behavior
between FRP bars and concrete under thermal loads based on the bond-slip relationship and
stated that effects of elevated temperatures are much more significant when the concrete cover
thickness is small. Ünlüoğlu et al. (2007) experimentally investigated the effects of concrete

F
cover on material properties of rebars at elevated temperatures. They presented the results to

OO
determine the most appropriate amount of concrete cover and emphasized that concrete cover
is a big precaution regarding the structural performance of RC members against fire. An
experimental study was carried out by Karanfil (2007) with four different concrete cover
thicknesses frequently used in structures in order to estimate the damage that may occur in
PR
rebars as a result of the fire effect. The conclusions show that the concrete cover has a great
influence on protecting the rebars against adverse effects of elevated temperatures. Wang et al.
(2007) experimentally studied the mechanical properties of CFRP and GFRP bars after high
temperatures. Test results show that the stress-strain relationship for FRP bars at high
OR

temperatures is almost linear until fracture moment. Bingöl and Gül (2009b) investigated the
bond strength between concrete and rebar after high temperatures. Losses in concrete
compressive strength and concrete-rebar bonding with increasing temperature level were
TH

determined. It was stated that the losses are more in water-cooled samples, which may be due
to the thermal shock resulting from sudden temperature changes.
AU

Masmoudi et al. (2010) investigated thermal effects on GFRP bars over pull-out tests. It has
been found that up to 60°C, the loss in bond strength is not significant. Alongside the
determination of bond strength, bond-slip relationship was modeled in the study using the
CMR-model. Robert and Benmokrane (2010) evaluated the variation of mechanical properties
of sand-coated GFRP reinforcing bars subjected to low and elevated temperatures. Tensile,
shear and flexural properties were investigated to get an overview on the thermal stability of
mechanical properties of GFRP bars subjected to large variations of temperatures. It was stated
that at very high temperatures, as close to the glass-transition temperatures of the polymer
matrix, mechanical properties, especially stiffness and strength of the composites decreases
considerably. Burnaz (2010) developed fire behavior models suitable for fire-affected concrete
and rebars, and investigated the structural behavior of RC members at elevated temperatures by
6
using computational nonlinear analyzes. Within the research, comparative examples on the
effect of construction material types were demonstrated over previous research. Alsayed et al.
(2012) investigated the residual tensile properties of GFRP bars after being subjected to
elevated temperatures for different periods. Alongside the losses in tensile strength,
proportional to the level of temperature and exposure period; the bars with concrete cover
showed higher residual tensile strength compared to their counterparts which were exposed to
direct heating. Carvelli et al. (2013) performed an experimental investigation on the static
behavior of concrete beams reinforced with GFRP bars exposed to localized elevated
temperatures, by considering maximum temperature and lapping scheme of the bars. The results

F
show that geometry of the reinforcement has a more relevant influence on the ultimate load

OO
than on the initial stiffness of the specimens. Pagani et al. (2014) presented a methodology to
assess the dangerousness of the bushfire event on bridge decks reinforced with GFRP bars.
Analytical and FEM simulations were performed on a strip representative of the bridge slab
behavior in terms of both deformation and failure mode, and some suggestions regarding
PR
retrofitting or concrete cover increase were made. Hamad et al. (2017) performed an
experimental study on the effect of elevated temperatures on the mechanical properties of FRP
bars and the bond behavior between FRP bars and concrete. Four types of rebars, which are
BFRP, CFRP, GFRP, and steel bars were used and exposed to high temperatures. An empirical
OR

model was also proposed to predict the post-heating bond stress-slip relationship between FRP
bars and surrounding concrete. Yang et al. (2017) presented the results of the microscopic
structures through SEM, FTIR, DSC, and tensile properties of GFRP bars, which were
TH

employed to investigate the combined effects of curing temperatures and alkaline concrete on
tensile properties of GFRP bars. The results show that the influence of different curing
temperatures on the tensile strength of GFRP bars is different between the bare bar and the bars
in concrete. Ashrafi et al. (2017) studied the effects of physical and thermal properties of
AU

various FRP bars on their performance at elevated temperatures considering the parameters as
bar diameter, fiber type, resin type, fiber to matrix ratio, and thermal properties. The results
show that all of these parameters except the fiber to matrix ratio are effective on the tensile
behavior of FRP bars.

7
3. EXPERIMENTAL PROGRAM AND RESULTS

In order to attain characteristics of concrete compressive behavior, tensile behavior of steel and
GFRP bars, and bonding behavior of these bars; ready-mixed concrete of concrete with 40 MPa
was ordered while yield strength of steel bars is 500 MPa and ultimate strength of sand-coated
GFRP bars is 900 MPa, referring to the manufacturer’s identification documents. Mentioned
material characteristics have been experimentally obtained at room temperature as 43.4 MPa,
505 MPa and 918 MPa, respectively. GFRP bars in this study were manufactured with ratios of
80% glass fiber and 20% epoxy resin (by weight) and surface coating of the bars contains

F
washed sea sand of 0.3-0.5 mm diameters. Filament diameter, tensile strength and tensile
modulus of glass fibers are 24 μm, 2700 MPa and 81.2 GPa, respectively. Concrete cubes, bare

OO
steel and GFRP bars, and pull-out specimens were heated from 100°C to 800°C in an electrical
high-temperature oven (Figure 1) for an hour. It should be noted that an electrical based oven
cannot reflect real situations of fire. For instance, Li et al. (2004) and Zhao and Sanjayan (2010)
compared the fire effects with different kind of ovens (electrical, gas, oil based) and reported
PR
that heating effects on the structural properties of materials would vary especially regarding the
divergent temperature elevation speed of ovens. Before the heating process of pull-out
specimens, thermal insulating coat was applied on the outer parts of steel and GFRP bars, which
were not embedded in concrete. Because bond strength of these bars are intended to be found,
OR

it is important to maintain same conditions on all of the specimens and to provide a heat
protection.
TH
AU

Figure 1. Pull-out specimens in the electrical high-temperature oven.

8
3.1. Compressive Strength Tests of Concrete

Throughout studying the effects of elevated temperatures, concrete has the most important role
in RC structural members. In the case of elevated temperature effect, it could be remarked that
concrete indirectly affects the mechanical properties of reinforcement bars it contains. This
indirect effect could be summarized as the heat conduction to the rebar through the concrete and
degradation of the bonding performance owing to the deterioration of the concrete. Higher
performance of concrete would protect the rebar from outer temperature and transfer the loads to
the rebar in a more efficient manner. In order to see structural changes in the concrete at elevated

F
temperature levels, three samples of 15×15×15 cm3 concrete cubes were allowed to stand for 1

OO
hour in each temperature group and left to cooling at room temperature. According to the results
of tests carried out in the direction of ASTM C39/C39M-17b (2017), with the loading rate of 0.25
MPa/s, decrease curve of the concrete compressive strength related with the target temperatures
is demonstrated in Figure 2. Although ASTM C39/C39M-17b (2017) recommends cylindrical
PR
concrete specimens for compressive strength test, authors preferred cubic concrete specimens in
order to provide a better concordance with pull-out test specimens.
OR
TH
AU

Figure 2. Compressive strength variation of concrete cubes subjected to target temperatures

Since the cement and aggregate forming the concrete, contain silica and limestone, it is expected
that strength loss depends on various parameters. Particularly the quartz in silica-based
coarse/fine aggregates, is subjected to polymorphic change at 570°C temperature, and transforms
from α quartz into β quartz. This transformation causes volume increase and damage in concrete
(Yüzer et al. 2001; Hyden et al. 1965). Besides, in calcareous and dolomitic aggregates, carbonate

9
transforms into CaO or MgO at 800-900°C. As temperature increases, limestone or dolomite
expands; decomposition of CO2 and formation of CaO or MgO initiate shrinkage. Those volume
changes also cause damage in concrete (Yüzer et al. 2001; Khoury 1992). According to the results
of this study, 45% of the concrete compressive strength at 600ºC is preserved, but residual
strength is only 18% at 800ºC. These outcomes demonstrate the higher material performance
level of concrete used today at elevated temperatures compared to previous studies.

Eurocode 2 Part 1-2 (2004) suggests that reduced concrete compressive strength could be
assumed 60% at 500ºC, 45% at 600ºC and 15% at 800ºC compared to the initial value, which

F
are compatible with the results of this study (500ºC: 58%; 600ºC: 45%; 800ºC: 18%). Regarding

OO
the tensile strength of concrete, Eurocode 2 Part 1-2 (2004) suggests a linear degradation to
assume strength value as zero at 600ºC, which aims to lead safer designs. However,
experimental studies reveal divergent results. For instance, Zhang et al. (2002) presented that
concrete compressive strength is 54% and concrete tensile strength is 24% at 600ºC with respect
PR
to the initial values. Although tensile strength tests of concrete have not been performed in this
study, it is possible to assume that degradation of tensile strength of concrete is faster than
compressive strength degradation.
OR

3.2. Tensile Strength Tests of Steel and GFRP bars

In order to provide mechanical properties of steel and GFRP bars under elevated temperature
TH

conditions; except the control group (23ºC), all specimens have been heated for 1 hour at
elevated temperatures between 100ºC-800ºC and then left to cooling at room temperature.
Entire length of the bars have been exposed to elevated temperatures without any protection.
AU

Axial tensile tests of three bars were performed in each temperature group as compatible with
ASTM A370-17 (2017), ASTM D7205/D7205M-06 (2016) and ACI 440.3R-12 (2012)
recommendations, with the loading rate of 2 mm/min (Figure 3). Because claws of the testing
machine have the possibility to damage GFRP bars, a protection using steel tubes was applied
after heating stage of those rebars in order to acquire accurate results. Steel tubes have an
outside diameter of 35 mm, nominal wall thickness of 4.5 mm and length of 300 mm. Two steel
tubes were bonded to the free ends of GFRP bars using epoxy adhesive. This epoxy has a
bonding strength of 3.5 MPa and density of 1.70 kg/L after 7 days of curing. For steel (db=12
mm) and GFRP (db=9 mm) bars in each temperature group, variation of the mean values of
axial tensile strength and ultimate strain are summarized in Figures 4 and 5, while stress-strain
curves of the bars are presented in Figures 6 and 7.

10
 F
OO
PR
Figure 3. Axial tensile strength tests of steel and GFRP bars
OR
TH
AU

Figure 4. Tensile strength variation of steel and GFRP bars subjected to target temperatures

11
 F
OO
Figure 5. Ultimate strain variation of steel and GFRP bars subjected to target temperatures

PR
OR
TH

Figure 6. Typical stress-strain curves of steel bars subjected to target temperatures

AU

Figure 7. Typical stress-strain curves of GFRP bars subjected to target temperatures

12
It is a known fact that yield/ultimate strength and elasticity modulus values of steel bars
decrease at high temperatures. Milke (1988) reported that typical construction steel is
crystallized at 650°C. According to Lie (1992) and SFPE (2002), it was stated that steel
generally maintains 50% of tensile strength and stiffness under environmental conditions with
a temperature of 593°C (1000°F) same as concrete. Decrease value of those is 80% at 700°C
and almost 100% at 1200°C (Karanfil 2007; Gewain 2003). Stress-strain curves of steel bars
(Figure 6) in the range of 23-600°C almost overlap each other and represents typical tensile
behavior at room temperature. Heating level of 600°C only increases the average strain value
by 1%. However, it is seen that exposure to 800°C results in decrease of yield and ultimate

F
strength alongside increase of ultimate strain, which corresponds to an excessive increment of

OO
the ductility although elastic behavior did not considerably change. Additionally, even it is not
presented in the figures, yield strain values of steel were 2.30-2.40‰ between 23°C and 600°C,
while it is 1.73‰ at 800°C. Accordingly, Young’s modulus of steel bars remained nearly
constant (≈210 GPa) at all temperature levels. As it is seen throughout the test results, yield
PR
strength of steel bars does not change up to 600°C; almost 30% of decrease on the yield strength
and 75% of increase on the ultimate strain occurs at 800°C alongside the diffraction of outer
layer. Those results, which are incompatible on tensile strength with previous studies and
general acceptance, show that material properties of steel has been improved over the years and
OR

present characteristics of steel allows to be used even at 800°C.

For the purpose of explaining the diffraction of outer layer of steel bars at 800°C, it should be
TH

mentioned that three main zones are observed in a typical cross-section of the bars, which are:
(1) a ferrite-pearlite core; (2) an intermediate layer with a mixture of bainite and ferrite; and (3)
a tempered martensite layer on the surface. Microstructural changes are mostly concentrated in
AU

the outer layer of the steel bars during heating and extensive softening of the outer hard layer
generally occurs after 600°C. That temperature point is dependent to the different chemical
composition and has a range of 600-800°C (Ergün et al. 2009; Nikolaou and Papadimitriou
2004).

The structure of the polymer matrix in FRP bars is the most important determinant in terms of
the fire safety. Deterioration of the mechanical properties of polymer matrix due to elevated
temperatures, results in a decrease in the mechanical and bonding properties of the FRP
material. This is because the matrix forms the basis of the composite and keeps the fibers
together to increase the strength of the composite. Neither vinylester nor unsaturated polyester-

13
styrene system is suitable for high temperature applications (Yavuz 2011). Regarding GFRP
bars, previous studies suggest lower temperature levels for the safety of the structural member.
Those studies report that polymeric matrix of GFRP bars starts to lose their strength after 70ºC
and disintegrates from the fibers; however, an epoxy resin can be used instead to provide a
resistance up to 160ºC. Despite above statements, ultimate tensile strength values of GFRP bars
decrease gradually as the temperature increases until 300°C, however, it was seen in this study
that most of the strength is maintained. After 300°C, sand-coating of the bars started to thaw
and ultimate strength value dropped from 918 MPa (23°C) to 330 MPa at 400°C, and to 94 MPa
at 500°C. Ultimate strain values of GFRP bars increases slightly as the temperature increases,

F
however, increase rate jumps after 300°C. At 600°C and 800°C, it is also important to note that

OO
sand-coating and matrix of the GFRP bars melted away completely, and glass fibers within the
bars scattered. Hence, axial tensile tests of the GFRP bars at those temperature levels could not
be performed. Stress-strain relationship of GFRP bars (Figure 7) also demonstrates the
mentioned tensile behavior diversity. As it is seen in the figure, rupture of GFRP bars in the
PR
range of 23-400°C were sudden. Although any apparent physical change was not observed prior
to the tensile tests, GFRP bars subjected to 500°C exhibited a nonlinear tensile behavior until
ultimate strength point and after then, an exceptional rupture behavior. This outcome reveals
that epoxy resin in the bars suffered damage during heating and bonding failure of the glass
OR

fibers prevented the simultaneous loading on all of the fibers. Additionally, Young’s modulus
of GFRP bars demonstrated a similar gradual degradation from 23°C to 500°C. Corresponding
values were calculated as 49.89, 47.04, 43.73, 41.68, 37.78, 32.56, 13.54 and 2.80 GPa,
TH

respectively.

3.3. Pull-out Tests of Steel and GFRP bars

AU

Steel (db=12 mm) and GFRP (db=9 mm) bars were subjected to pull-out tests throughout an
identical testing procedure in accordance with the recommendations of ASTM A944-10 (2015),
ASTM D7913/D7913M-14 (2014) and ACI 440.3R-12 (2012) in order to perform a more
reliable comparison on the bond strength characteristics at elevated temperatures (Figure 8). A
displacement controlled pull-out test procedure was performed for each of the bars with the
loading rate of 1 mm/min. Although related codes recommend testing with free-ends at top and
bottom faces of concrete cube, this study aims to investigate the bond strength of rebars in order
to simulate the real bonding behavior in RC members. Hence, steel and GFRP bars were kept
in the concrete by leaving a 25 mm cover at the bottom face of the concrete cube. At a distance

14
of 5 times the diameter of the rebar, steel and GFRP bars were embedded in the concrete cube
and the remaining part of the bars was left in a plastic tube for the purpose of attaining a
proportional constant embedment length for all the bars. Hence, embedment length of steel bars
is 60 mm while that of GFRP bars is 45 mm.

F
OO
PR
OR
TH

Figure 8. Pull-out testing scheme of steel and GFRP bars

Application of a heat insulation method using fire plaster was preferred in order to attain
resistance to high temperatures for steel and GFRP bars. This fire plaster is in A1 class of fire
AU

resistance and T1 class of thermal conductivity. Free-ends of the pull-out specimens were
completely covered with fire plaster before heating process (Figure 9). After then, fire plaster
was removed and steel specimens were directly subjected to the testing, while free-ends of
GFRP specimens were bonded with epoxy adhesive and embedded in steel tubes in order to
prevent physical damage on the bars by machine claws.

15
 (a) (b)

F
OO
Figure 9. Pull-out specimens of rebars: (a) covered with fire plaster; (b) fire plaster was
PR
removed after heating process

Since embedded parts of the bars were not projected from the bottom face of concrete cubes,
OR

axial displacement values of the bars were obtained from the loaded-end of the bars via internal
transducers of the testing machine. Bond failure load of steel bars are less than half of the yield
strength, thus, strain of steel pull-out specimens were limited in the elastic region. Elongation
of steel bars at each load level was calculated considering loaded-end length and the length in
TH

plastic tube in order to obtain slippage values by subtracting the elongation of bars from total
displacement. Regarding GFRP pull-out specimens, since free-ends of the bars were covered
with high-strength epoxy adhesive and bonded with steel tubes, elongation of the bars were
AU

calculated considering only the length in plastic tube, and slippage values were obtained
similarly as following:

𝐹∙𝐿𝑓
𝑆𝑒 = (1)
𝐸∙𝐴

𝑆𝑠 = 𝑆𝑡 − 𝑆𝑒 (2)

where Se is elastic elongation of the bar; F is applied maximum force; Lf is total free length of
the bar (Lc=335 mm for steel bars, Lc=130 mm for GFRP bars); E is Young’s modulus of the
bar; A is cross-sectional area of the bar; Ss is clear slip value; and St is total axial displacement.

16
Bond strength values of the bars (τ) were calculated in terms of the surface stress as the
proportion of the applied force to the embedded surface area of the bars as following:

𝐹
𝜏= (3)
𝐶∙𝐿𝑏

where C is equivalent circumference of the bar; and Lb is bonded length (Lb=60 mm for steel
bars, Lb=450 mm for GFRP bars). Bond strength between concrete and rebar was measured
regarding the high temperature behavior of the RC members following the test procedure as
presented in Figure 10. Same as the tensile test specimens, in order to prevent claw damage on

F
GFRP bars, a steel tube was bonded to the free outside end of the GFRP bars using epoxy

OO
adhesive after the heating process.

PR
OR
TH
AU

Figure 10. Pull-out tests of steel and GFRP bars

For steel (db=12 mm) and GFRP (db=9 mm) bars in each temperature group, variation of the
mean values of bond strength and slippage are summarized in Figure 11; bond strength-slip
curves of the bars at each temperature group are presented in Figures 12 and 13; and all of the
material characteristics are summarized in Table 1 in the form of a scaled representation
considering the related characteristic value at 23ºC. Observation of the pull-out specimens after

17
testing revealed that no apparent cracks exist on the top face of concrete cubes and there was
not any apparent damage on the bars. Neither splitting of the concrete nor peeling of the bar
occurred as the cause of slippage; therefore, bond failure mechanism depends upon the shear
failure in concrete between the ribs of steel bars and sandy ribs on the coating of GFRP bars.
Surface deterioration of GFRP bars also accelerates bond strength degradation. Tensile rupture
load values of steel and GFRP bars are more than twice of slippage load values of the bars ,
thus, tensile failure of the bars did not precede bond failure. For instance, yield load-slippage
load matching for steel bars demonstrates 57.1 kN-28.9 kN at 23ºC and 40.3 kN-3.4 kN at
800ºC. Similar rupture load-slippage load matching for GFRP bars reveals 58.4 kN-12.7 kN at

F
23ºC and 6.0 kN-5.7 kN at 500ºC. Besides, it is important to emphasize that free-ends of the

OO
bars were covered with fire plaster in order to attain heat insulation and mentioned tensile
strength levels are higher at those temperatures. Bars are assumed to be protected from elevated
temperature effects owing to the heat resistance of the covered fire plaster. It is also possible to
mention that the bar section, which was not covered by fire plaster and left in the plastic tube,
PR
was subjected to heat conduction through the concrete. Despite this situation might affect the
tensile characteristics of the bars, it was not feasible to measure that effect.
OR
TH
AU

Figure 11. Bond strength variation and slippage values of steel and GFRP pull-out specimens
subjected to target temperatures

18
 F
OO
Figure 12. Typical bond strength-slip curves of steel bars subjected to target temperatures

PR
OR
TH

Figure 13. Typical bond strength-slip curves of GFRP bars subjected to target temperatures
AU

Table 1. Scaled variation of material characteristics at elevated temperatures

Target Temperatures (°C)

23 100 150 200 250 300 400 500 600 800

Concrete Compressive Strength 1 0.99 0.92 0.85 0.80 0.77 0.69 0.58 0.45 0.18
Material Characteristics

Steel Yield Strength 1 0.98 0.98 0.99 0.97 0.98 0.98 0.98 0.97 0.70

GFRP Ultimate Strength 1 0.96 0.91 0.86 0.79 0.70 0.36 0.10 - -

Steel Bond Strength 1 0.92 0.87 0.85 0.77 0.75 0.74 0.65 0.59 0.12
GFRP Bond Strength 1 0.82 0.77 0.77 0.69 0.62 0.57 0.45 0.34 -

Steel Bond Slip 1 0.99 1.02 1.01 0.99 0.99 0.97 0.85 0.80 0.68
GFRP Bond Slip 1 0.86 0.76 0.77 0.74 0.74 0.72 0.62 0.57 -

19
Pull-out testing results demonstrate that a gradual decrease appears for the bond strength of
both steel and sand-coated GFRP bars, with the temperature increase. It is seen that 74% of the
bond strength is maintained at around 400ºC for steel bars; it drops to 59% at 600ºC; and almost
90% of bond strength has been lost at 800ºC. A previous experiment study by Katz and Berman
(2000) reported that GFRP bars lost most of the bond strength and maintained only 10-20% of
the initial peak values after 200°C. Nevertheless, it has been observed in this experimental study
that GFRP bars maintain 77% of their maximum bond strength at 200ºC, while their bond
strength is almost 35% of the initial value at 600ºC. Concrete pull-out specimens of GFRP bars
exploded during heating at 800°C, therefore, the results do not include the bond strength at that

F
temperature. These results reveal that fire performance of GFRP bars is higher compared to

OO
those in the previous studies, because of especially the concrete cover of 25 mm in this study,
alongside the developments in production process of the material. Because standard testing
procedure suggests projecting bars from the bottom face of the concrete cube, bars are subjected
to direct effect of elevated temperatures and this effect on the surrounding concrete around the
PR
embedded bar section is more significant. Steel bars preserve their initial slippage value until
400ºC, while GFRP bars do not show a significant decrease until 500ºC. Although steel bars
have higher slippage values, the decrease for both of the materials at higher temperatures are
not significant overall with respect to their bond strength variation.
OR

Regarding bond strength-slip relationship of the tested bars, both of the material types represent
almost linear ascending and descending curve branches in general. Slope of the ascending
TH

branch decreases by target temperature for steel bars, while any significant change could not be
determined for GFRP bars. Descending branch of the curves for both of the materials also
demonstrates the failure mode. Because the bond strength between concrete and bar is lower
AU

than the splitting tensile strength of concrete, slippage of the bars depends upon the fact that
concrete between the ribs was sheared off. Slopes of the descending branch for steel bars are
close to each other while gradual reduction of the slope could be seen for GFRP bars as target
temperature increases. Since polymeric material of GFRP bars has higher transverse coefficient
of thermal expansion than concrete, bonding performance would be reduced and surrounding
concrete could explode at elevated temperatures. Deterioration of the integrity of surface
coating would result in a more reduction of the bonding performance. Considering that pull-out
specimens of this study have exploded at 800ºC, irregular bond strength-slip behavior of the
GFRP specimens at 600ºC could be correlated with these issues. In addition, slightly fluctuant
appearance of the bond strength-slip curves of steel and GFRP bars between 23-400ºC is also
remarkable that bonding performance transforms as the temperature increases.
20
Scaled variation of material characteristics in Table 1 reveals the relationship of material
characteristics at elevated temperatures. It is initially important to respecify that bond strength
degradation does not depend on tensile strength of the bars. Pull-out testing has been performed
by keeping the bars inside the concrete cube with a concrete cover of 25 mm and covering free-
ends of the bars with fire plaster. Considering that concrete cover would not even totally isolate
the bars from outer temperature, tensile strength of the bars did not descend under the bond
strength level. For instance, Ünlüoğlu et al. (2007) reported that 25 mm of concrete cover
provides nearly 250ºC decrease of outer temperature in terms of the heat conducted to the
embedded steel bars in concrete. Test results demonstrate that bond strength degradation is

F
mainly dependent upon the concrete compressive strength degradation. While bond strength

OO
variation of steel bars is mostly in correlation with concrete compressive strength variation,
bond strength degradation of GFRP bars is seen to be more apparent. This remarks that surface
deterioration of the bars is the secondary reason since coating of GFRP bars is affected much
more than surface of steel bars. Even any apparent damage on the surface of GFRP bars has not
PR
been observed, bond slip variation of GFRP bars supports this evaluation.

4. EMPIRICAL MODELING OF REBAR BOND STRENGTH AT ELEVATED

TEMPERATURES
OR

An empirical model for the bond strength of rebars, which are subjected to elevated
temperatures, is developed based on the experimental results of this study. Regarding further
TH

application of the model, it should be noted that the model estimates the behavior of rebars with
a concrete cover of 25 mm. Since increasing the concrete cover would enhance the bond
strength, this effect could be considered using another coefficient following adequate
AU

experimental investigation. All of the parameters in the model are derived from actual material
properties and adapted to following stages in order to attain a simplified model. Considering
that concrete cubes of GFRP pull-out specimens exploded at 800ºC, also bond strength value
of steel rebars is out of the general variance behavior at that temperature, proposed model has
the limits of 23ºC and 600ºC. Initially, as a consistent empirical modeling pattern, Eq. (4) that
takes into account the bonding performance of rebars under elevated temperatures is described.

𝜏𝑚,𝑇 = 𝛾𝑚 ∙ 𝜂𝑚 ∙ 𝑓 (𝑇) , 23 ≤ 𝑇 ≤ 600 (4)

where 𝜏𝑚,𝑇 is residual bond strength value of the material subjected to temperature of TºC; 𝛾𝑚
is material bond strength coefficient; 𝜂𝑚 is material heat vulnerability coefficient; and 𝑓 (𝑇) is
21
heat behavior function. Because FRP rebars could have very different characteristics in terms
of fiber material or surface coating, also bond strength variation mainly depends on the
characteristics of concrete; bonding behavior of steel rebars have been placed in the basis of
this modeling approach. Material bond strength coefficient (𝛾𝑚 ), is actually the proportion of
the maximum bond strength value (at 23ºC) of the material m to that of the steel. Hence, bond
strength coefficient of steel (𝛾𝑠 ) is considered as 1.000, while:

𝜏𝑔,23
𝛾𝑔 = 𝜏𝑠,23
(5)

F
where 𝛾𝑔 is bond strength coefficient of GFRP; while 𝜏𝑔,23 and 𝜏𝑠,23 are initial bond strength

OO
values of GFRP and steel bars at room temperature (23ºC), respectively. Since bond strength
degradation of steel and GFRP bars have almost linear characteristics, second and third
components of Eq. (4) were constituted using curve fitting and mathematical association
processes. Considering that the reference material of the empirical formulation is the steel rebar,
PR
heat vulnerability coefficient of steel (𝜂𝑠 ) is 1.000, while that of the GFRP is calculated
regarding the following simplified equation:

1473−𝑇
OR

𝜂𝑔 = (6)
1450

Finally, heat behavior function is formulated as following, taking into account the bond strength
of steel at room temperature, 𝜏𝑠,23 , and the effect of temperature, T, again using the relationship
TH

of the bond strength curves and mathematical simplification processes.

𝑇
𝑓 (𝑇) = 0.981 ∙ 𝜏𝑠,23 ∙ (1 − 1450 ) (7)
AU

Regarding Eqs. (4) through (7), parameters of the empirical model, predicted bond strength
values and comparison with the experimental data are presented in Table 2. Those predicted
values of steel and GFRP bars are obviously consistent with experimental data. However, it is
important to state that pull-out testing scheme in this study regards the usage of rebars in RC
members and a concrete cover of 25 mm is left at the end of the rebars. Therefore, direct effect
of the elevated temperatures into rebars is avoided. For example, some previous research
performed the pull-out testing process otherwise and bond strength values of FRP rebars are
much lower especially after 300ºC. However, empirical model of this study was applied to the
experimental data of those studies and it has been found that predicted bond strength of steel

22
bars is consistent while that of various types of FRP bars differs from the experimental bond
strength generally after 200ºC. This inconsistency depends on the fact that sand-coating of the
FRP bars starts to thaw and surrounding concrete is affected more near those temperature levels.

Table 2. Comparison of empirical and experimental bond strengths at elevated temperatures

Bond Strength Heat Vulnerability Bond Strength

Temp. Error
Material Coefficient Coefficient (MPa)
(ºC) (%)
𝛾𝑠 , 𝛾𝑔 𝜂𝑠 , 𝜂𝑔 𝜏𝑒𝑚𝑝 𝜏𝑒𝑥𝑝
23 12.32 12.76 3.43
100 11.65 11.77 1.00
150 11.22 11.05 1.55

F
200 10.79 10.81 0.15
Steel 250 1.000 1.000 10.36 9.81 5.58

OO
300 9.93 9.54 4.05
400 9.06 9.40 3.60
500 8.20 8.25 0.57
600 7.34 7.50 2.14
23 PR 1.000 9.60 9.94 3.48
100 0.947 8.60 8.18 5.04
150 0.912 7.98 7.69 3.76
200 0.878 7.38 7.67 3.79
GFRP 250 0.779 0.843 6.81 6.90 1.41
300 0.809 6.26 6.22 0.64
400 0.740 5.23 5.69 8.11
OR

500 0.671 4.29 4.47 4.03

600 0.602 3.44 3.38 1.70
Average Error 3.00
𝜏𝑒𝑚𝑝: Empirical bond strength, 𝜏𝑒𝑥𝑝: Experimental bond strength.
TH

Comparison of prediction errors on the bond strengths of previous research was presented by
percentage in Table 3. In advance, it should be noted that none of the cited research includes
identical pull-out testing process. Bingöl and Gül (2009b) and Ergün et al. (2016) performed
AU

pull-out tests of steel bars without a projection from concrete cylinders; while Robert and
Benmokrane (2010), Masmoudi et al. (2011), El-Gamal (2014), Hamad et al. (2017) and Lei et
al. (2018) preferred projecting the rebars forum the bottom face of the concrete cube. Although
bond strength variation of steel bars is compatible with the modeling approach, FRP bars (glass,
basalt and carbon) reveals excessive deviation after 200-300ºC due to the direct effect of heating
on the bars. Since only FRP pull-out tests were performed within some of those research,
empirical bond strength values have been calculated compulsorily referring to initial value of
FRP bars. Nevertheless, prediction errors are still promising. Being a significant research, Katz
and Berman (2000) performed a comparison between steel and different types of GFRP bars,
and pull-out tests were performed during the heating phase. Thus, proposed empirical model
could not be applied on those results and were not compared in this study.

23
 Table 3. Comparison of prediction errors on the bond strengths of previous research (%)

Bingöl and Robert and Ergün

Masmoudi El-Gamal Lei et al.
Gül Benmokrane et al. Hamad et al. (2017)
et al. (2011) (2014) (2018)
(2009b) (2010) (2016)
Temp.
Steel GFRP GFRP GFRP Steel Steel GFRP BFRP CFRP GFRP CFRP
(ºC)
23 -3.46 -3.05 -3.05 -3.72 -3.45 -3.46 -3.46 -3.46 -3.46 -3.05 -3.05
30 - - - - - - - - - -5.65 2.34
40 - -5.09 -4.51 - - - - - - - -
50 14.39 -4.50 - - - - - - - - -
60 - - -5.17 - - - - - - 2.48 10.96
80 - - 3.20 - - - - - - 11.06 *
100 -13.19 - - -1.57 -10.52 - - - - - -

F
125 - - - - - -5.16 14.80 6.63 -0.79 - -
150 -0.95 - - - - - - - - - -

OO
200 2.08 - - -13.06 -1.71 - - - - - -
250 -0.46 - - - - 1.35 37.50 75.22 21.49 - -
300 -7.24 - - -15.37 - - - - - - -
325 - - - - - 4.47 189.65 185.85 225.88 - -
350 -6.04 - - 19.94 - - - - - - -
375 - - - PR - - 5.05 * * 430.94 - -
400 6.21 - - - -14.04 - - - - - -
450 -11.63 - - - - - - - - - -
500 7.45 - - - - - - - - - -
600 14.02 - - - 6.57 - - - - - -
* Those values were not provided within the cited research.
OR

5. CONCLUSIONS

Theoretical-experimental based scientific studies have great importance in order to expand the
TH

usage areas of new construction materials and improve these materials. In this study, structural
performance of construction materials subjected to elevated temperature effects, were observed.
Compressive strength of concrete, tensile strength and bond strength of steel and GFRP bars
AU

were experimentally investigated. Comparison between the suggested mathematical model and
experimental results demonstrates a consistent behavior in terms of the bond strength of rebars.
That empirical modeling approach has the capacity of predicting rebar bond strength at elevated
temperatures, and could minimize the necessity of comprehensive experimental programs.

Investigation of the experimental results reveals that concrete compressive strength, steel yield
strength and GFRP ultimate strength exhibit degradation at high temperature levels owing to
the different material characteristics. Expansion and shrinkage of the components in cement
and aggregates, deterioration of the molecular structure of steel rebars, and dispersion of the
matrix holding the composite structure of GFRP rebars, due to elevated temperature effects, are

24
already known reasons of decrease tendency in terms of the structural performance. Despite the
use of silica based cement and limestone based aggregate in the concrete, concrete specimens
could preserve 45% of the initial compressive strength at 600ºC and 18% at 800ºC. When axial
tensile test results of steel and GFRP bars exposed to elevated temperature effects in
unprotected conditions are examined, there is no change in the yield strength of steel bars up to
600ºC, but only about 30% loss at 800ºC. GFRP bars, however, exhibit a gradual loss of strength
as the temperature level increases, but as dissimilarly with many of the previous studies,
presents a higher performance and there is a serious loss of strength only after 300ºC. After
500ºC, sand-coating and matrix of the GFRP bars melted away completely, and glass fibers

F
within the bars scattered.

OO
Variation of the bond-strength values of steel and GFRP bars with temperature level, reveals
similar and almost linear curves. Although concrete compressive strength decreases rapidly in
relation to the temperature, bond strength of both types of bars decreases at a slower rate. It is
PR
obvious that, on the basis of concrete-rebar bonding, sand-coating of the GFRP bars presents a
lower contribution compared to the ribs of the steel bars, and it is possible to improve this weakness
of the GFRP bars with various solutions. However, bond strength performance of GFRP bars,
which is close to that of the steel bars at normal conditions, and non-sudden decrease of the bond
OR

strength as the temperature increases, suggest that this material performs far better than expected,
against elevated temperature effects owing to the developments in production process of the
material. Enhancing the structural performance of RC members by protecting rebars against
TH

elevated temperature effects, could be achieved by especially maintaining an adequate concrete

cover and reducing the amount of silica in the cement compound. Various types of cement, in
which amount of silica content is much less and compressive strength is much better, are recently
AU

available as an alternative enhancement approach.

Additionally, empirical modeling approach in the study reveals a higher consistency considering
the error percentage on the experimental data. Maximum prediction error is 5.58% for steel bars,
while it is 8.11% for GFRP bars. This outcome suggests that the same modeling approach could
be performed on other types of rebar materials for predicting the bond strength without the
necessity of wide-range testing processes. For the purpose of a wide-ranging formulation, heat
resistance of the concrete and surface coatings of FRP bars should be taken into consideration as
well as the concrete cover thickness, following many series of pull-out tests.

25
ACKNOWLEDGEMENTS

The authors would like to thank Europa Composite from İzmir, Turkey for providing GFRP
bars. This research was supported by Research Fund of the Erzincan University under grant
number FBA-2017-491.

REFERENCES

F
Abbasi, A., Hogg, P.J., 2005. Temperature and environmental effects on glass fiber rebar: modulus, strength and
interfacial bond strength with concrete. Composites Part B: Engineering, 36, 394-404.

OO
ACI 440.3R-12,2012. Guide Test Methods for Fiber-Reinforced Polymers (FRPs) for Reinforcing or
Strengthening Concrete Structures. American Concrete Institute, Michigan.

Alsayed, S., Al-Salloum, Y., Almusallam, T., El-Gamal, S., Aqel, M., 2012. Performance of glass fiber reinforced
polymer bars under elevated temperatures. Composites Part B: Engineering, 43, 2265–2271.
PR
Ashrafi, H., Bazli, M., Najafabadi, E.P., Oskouei, A.V., 2017. The effect of mechanical and thermal properties of
FRP bars on their tensile performance under elevated temperatures. Construction and Building Materials, 157,
1001-1010.
OR

ASTM Standard A370-17, 2017. Standard Test Methods and Definitions for Mechanical Testing of Steel Products.
ASTM International, West Conshohocken, PA.

ASTM Standard A944-10, 2015 Standard Test Method for Comparing Bond Strength of Steel Reinforcing Bars
to Concrete Using Beam-End Specimens. ASTM International, West Conshohocken, PA.
TH

ASTM Standard C39/C39M-17b, 2017. Standard Test Method for Compressive Strength of Cylindrical Concrete
Specimens. ASTM International, West Conshohocken, PA.

ASTM Standard D7205/D7205M-06, 2016. Standard Test Method for Tensile Properties of Fiber Reinforced
AU

Polymer Matrix Composite Bars. ASTM International, West Conshohocken, PA.

ASTM Standard D7913/D7913M-14, 2014. Standard Test Method for Bond Strength of Fiber-Reinforced Polymer
Matrix Composite Bars to Concrete by Pullout Testing. ASTM International, West Conshohocken, PA.

Banholzer, B., Brameshuber, W., Jung, W., 2006. Analytical evaluation of pull-out tests—the inverse problem.
Cement and Concrete Composites, 28, 564-571.

Benmokrane, B., Xu, H., Bellavance, E., 1996. Bond strength of cement grouted glass fiber reinforced plastic
(GFRP) anchor bolts. International Journal of Rock Mechanics and Mining Sciences & Geomechanics
Abstracts, 33(5), 455-465.

Bingöl, A.F., Gül, R., 2009a. Donatı-beton aderansı, yüksek sıcaklıkların beton dayanımına ve aderansa etkileri
konusunda bir derleme. TÜBAV Bilim Dergisi, 2(2), 211-230.

26
Bingöl A.F., Gül R., 2009b. Residual bond strength between steel bars and concrete after elevated temperatures.
Fire Safety Journal, 44(6), 854-859.

Burnaz, O., 2010. Betonarme Yapılarda Yangın ve Bu Yapıların Yangın Etkisi Altında Doğrusal Olmayan
Davranışlarının İncelenmesi. Ph.D. Thesis, Karadeniz Technical University, Trabzon.

Carvelli, V., Pisani, M.A., Poggi, C., 2013. High temperature effects on concrete members reinforced with GFRP
rebars. Composites Part B: Engineering, 54, 125-132

Chang, J.J., Yeih, W., Tsai, C.L., 2002. Enhancement of bond strength for epoxy-coated rebar using river sand.
Construction and Building Materials, 16, 465-472.

Chen, B., Liu, J., 2004. Residual strength of hybrid-fiber-reinforced high-strength concrete after exposure to high
temperatures. Cement and Concrete Research, 34, 1065-1069.

F
Cheng, A., Huang, R., Wu, J.K. and Chen, C.H., 2005. Effect of rebar coating on corrosion resistance and bond

OO
strength of reinforced concrete. Construction and Building Materials, 19, 404-412.

Chiang, C.H., Tsai, C.L., Kan, Y.C., 2000. Acoustic inspection of bond strength of steel-reinforced mortar after
exposure to elevated temperatures. Ultrasonics, 38, 534-536.
PR
Cullazoğlu, F., 2014. FRP Donatılı Betonarme Kirişlerin Eğilme ve Kesme Etkisi Altındaki Analizi. M.Sc. Thesis,
Niğde University, Niğde.

De Lorenzis, L., Rizzo, A., La Tegola, A., 2002. A modified pull-out test for bond of near-surface mounted FRP
rods in concrete. Composites Part B: Engineering, 33, 589-603.
OR

Diederichs, U., Schneider, U., 1981. Bond strength at high temperatures. Magazine of Concrete Research, 33(115),
75-84.

El-Gamal, S., 2014. Bond strength of glass fiber-reinforced polymer bars in concrete after exposure to elevated
temperatures. Journal of Reinforced Plastics and Composites, 33(23) 2151-2163.
TH

El-Hawary, M., Hamoush, S.A., 1996. Bond shear modulus of reinforced concrete at high temperatures.
Engineering Fracture Mechanics, 55, 991-999.

Ergün, A., Kürklü, G., Başpınar, M.S., 2009. Yüksek sıcaklık sonrası farklı sınıflardaki betonarme çeliklerinin
AU

mekanik özelliklerinin incelenmesi. Afyon Kocatepe University Journal of Science and Engineering, 9(2), 97-
103.

Ergün, A., Kürklü, G., Başpınar, M.S., 2016. The effects of material properties on bond strength between
reinforcing bar and concrete exposed to high temperature. Construction and Building Materials, 112, 691-698.

Eurocode 2, 2004. Design of Concrete Structures - Part 1–2: General Rules – Structural Fire Design (EN 1992-1-
2), European Committee for Standardization, Brussels.

Fava, G., Carvelli, V., Pisani, M.A., 2016. Remarks on bond of GFRP rebars and concrete. Composites Part B:
Engineering, 93, 210-220.

Fu, X., Chung, D.D.L., 1997. Improving the bond strength between steel rebar and concrete by increasing the
water/cement ratio. Cement and Concrete Research, 27(12), 1805-1809.

27
Galati, N., Nanni, A., Dharani, L.R., Focacci, F., Aiello, M.A., 2006. Thermal effects on bond between FRP rebars
and concrete. Composites Part A: Applied Science and Manufacturing, 37, 1223-1230.

Gallego, A., Jiménez, V., Gil, J., Piqueras, J. L., de Mata Vico, J., Rodríguez, J., 2004. Comparison between
concrete-black steel and concrete-galvanized steel bond via the pull-out test supplied with acoustic emission.
Proceedings EWGAE, 761-767.

Gewain, R.G., Iwankiw, N.R., Alfawakhiri, F., 2003. Facts for Steel Buildings Number 1 – Fire. American Institute
of Steel Construction (AISC), Chicago, IL.

Gustaferro, A.H., Abrams, M.S., Litvin, A., 1971. Fire resistance of lightweight insulating concrete. Lightweight
Concrete, ACI Publication SP-29, 161-180.

Haddad, R.H., Abendeh, R., 2004. Effect of thermal cycling on bond between reinforcement and fiber reinforced

F
concrete. Cement and Concrete Composites, 26, 743-752.

OO
Hamad, R.J.A., Johar, M.A.M., Haddad, R.H., 2017. Mechanical properties and bond characteristics of different
fiber reinforced polymer rebars at elevated temperatures. Construction and Building Materials, 142, 521–535.

Handoo, S.K., Agarwal, S., Agarwal, S.K., 2002. Physicochemical, mineralogical, and morphological
characteristics of concrete exposed to elevated temperatures. Cement and Concrete Research, 32, 1009–1018.
PR
Hashimoto, J., Takiguchi, K., 2004. Experimental study on pullout strength of anchor bolt with an embedment
depth of 30 mm in concrete under high temperature. Nuclear Engineering and Design, 229, 151-163.

Hayden, H.W., Moffatt, W.G., Wulff, J., 1965. The Structure and Properties of Materials, vol. 3. John Wiley &
OR

Sons, Inc., USA.

Jendele, L., Cervenka, J., 2006. Finite element modelling of reinforcement with bond. Computers and Structures,
84(28), 1780-1791.

Karanfil, H., 2007. Yüksek Sıcaklık Etkisi Altında Kalan Betonarme Yapılarda Çelik Donatı Özelliklerinin
TH

Değişimine Pas Payı Kalınlığının Etkisi. M.Sc. Thesis, Eskişehir Osmangazi University, Eskişehir.

Katz, A., Berman, N., Bank, L.C., 1999. Effect of high temperature on bond strength of FRP rebars. Journal of
Composites for Construction, 3(2), 73-81.
AU

Katz, A., Berman, N., 2000. Modeling the effect of high temperature on the bond of FRP reinforcing bars to
concrete. Cement and Concrete Composites, 22, 433-443.

Khoury, G.A., 1992. Compressive strength of concrete at high temperatures: Reassessment. Magazine of Concrete
Research, 44(161), 291-309.

Khoury, G.A., 1996. Performance of Heated Concrete - Mechanical Properties. Contract NUC/56/3604A with
Nuclear Installations Inspectorate, Imperial College, London, UK.

Larrard, F., Schaller, I., Fuchs, J., 1993. Effect of bar diameter on the bond strength of passive reinforcement in
high-performance concrete. ACI Materials Journal, 90(4), 333-339.

Lau, A., Anson, M., 2006. Effect of high temperatures on high performance steel fibre reinforced concrete. Cement
and Concrete Research, 36, 1698-1707.

28
Lee, H., Noguchi, T., Tomosawa, F., 2002. Evaluation of the bond properties between concrete and reinforcement
as a function of the degree of reinforcement corrosion. Cement and Concrete Research, 32, 1313-1318.

Lei, W., Yadon, M., Haibo, L., Shuang, C., Wei, L., 2018. Bond properties between FRP bars and coral concrete
under seawater conditions at 30, 60, and 80 °C. Construction and Building Materials, 162, 442-449.

Li M., Qian, C.X., Sun W., 2004. Mechanical properties of high-strength concrete after fire. Cement and Concrete
Research, 34(6), 1001-1005.

Lie, T.T., 2002. Structural Fire Protection. American Society of Civil Engineers, New York.

Masmoudi, R., Masmoudi, A., Daoud, A., Ouezdou, M.B., 2010. Thermal effects on bond properties of GFRP
rebars embedded in concrete. Journal of Civil Engineering and Architecture, 4(3), 1-5.

F
Masmoudi, R., Masmoudi, A., Ouezdou, M.B., Daoud, A., 2011. Long-term bond performance of GFRP bars in
concrete under temperature ranging from 20 °C to 80 °C. Construction and Building Materials, 25, 486-493.

OO
Milke, J., 1988. Analytical Methods for Determining Fire Resistance of Steel Members. SFPE Handbook of Fire
Protection Engineering, pp. 3/88-3/1 12, Society of Fire Protection Engineers, Quincy, MA.

Morley, P.D., Royles, R., 1980. The influence of high temperature on the bond in reinforced concrete. Fire Safety
PR
Journal, 2(4), 243-255.

Nikolaou, J., Papadimitriou, G.D., 2004. Microstructures and mechanical properties after heating of reinforcing
500 MPa class weldable steels produced by various processes (Tempcore, microalloyed with vanadium and
work-hardened). Construction and Building Materials, 18, 243–254.
OR

Pagani, R, Bocciarelli, M., Carvelli, V., Pisani, M.A., 2014. Modelling high temperature effects on bridge slabs
reinforced with GFRP rebars. Engineering Structures, 81, 318-326.

Robert, M., Benmokrane, B., 2010. Behavior of GFRP reinforcing bars subjected to extreme temperatures. Journal
of Composites for Construction, 14(4), 353-360.
TH

Rostasy, F.S., Weiss, R., Wiedemann, G., 1980. Changes of pore structure of cement mortars due to temperature.
Cement and Concrete Research, 10, 157-164.

SFPE (The Society of Fire Protection Engineers), 2002. SFPE Hand Book of Fire Protection Engineering, 3rd
AU

edition, Philip J. DiNenno (ed). National Fire Protection Association, Inc., Quincy, MA.

Tighiouart, B., Benmokrane, B., Gao, D., 1998. Investigation of bond in concrete member with fibre reinforced
polymer (FRP) bars. Construction and Building Materials, 12, 453-462.

Ünlüoğlu, E., Topçu, İ.B., Yalaman, B., 2007. Yüksek sıcaklıkta kalmış yapılarda pas payının betonarme çelik
donatı özeliklerine etkisi. İMO Teknik Dergi, 18(87), 4145-4155.

Wang, Y.C., Kodur, V., 2005. Variation of strength and stiffness of fibre reinforced polymer reinforcing bars with
temperature. Cement and Concrete Composites, 27, 864-874.

Wang, Y.C., Wong, P.M.H., Kodur, V., 2007. An experimental study of the mechanical properties of fibre
reinforced polymer (FRP) and steel reinforcing bars at elevated temperatures. Composite Structures, 80, 131-
140.

29
 Yamazaki, N., Yamazaki, M., Mochida, T., Mutoh, A., Miyashita, T., Ueda, M., 1995. Structural behavior of
reinforced concrete structures at high temperatures. Nuclear Engineering and Design, 156(1-2), 121-138.

Yang, W.R., He, X.J., Zhang, K., Yang, Y., Dai, L., 2017. Combined effects of curing temperatures and alkaline
concrete on tensile properties of GFRP bars. International Journal of Polymer Science, Article ID 4262703, 8
pages.

Yavuz, G., 2011. Lif takviyeli polimerlerin betonarme kirişlerde donatı olarak kullanımı. Engineering Sciences,
6(4), 1001-1015.

Yuan, G.L., Guo, C., Li, Q.T., Lu, Z.T., 2005. Bond damage in reinforced concrete caused by cooling after high
temperature. Zhongguo Kuangye Daxue Xuebao/Journal of China University of Mining and Technology, 34(5),
605-608.

F
Yüzer, N., Aköz, F., Öztürk, L.D., 2001. Yangına maruz yapılarda betonun basınç dayanımı-renk değişimi ilişkisi.

OO
Yıldız Teknik Üniversitesi Dergisi, 4, 51-60.

Zhang, B., Bicanic, N., Pearce, C.J., Phillips, D.V., 2002. Relationship between brittleness and moisture loss of
concrete exposed to high temperatures. Cement and Concrete Research, 32(3) 363-371.

Zhao, R., Sanjayan, J.G., 2010. Test method for concrete spalling using small electric furnace. Fire and Materials,
34, 189-201.
PR
Zoldners, N.G., Wilson, H.S., 1973. Effect of sustained and cyclic temperature exposures on lightweight concrete.
Behavior of concrete under temperature extremes, ACI Publication SP-39, 149-178.
OR
TH
AU

30

View publication stats